|Meta-analysis of oral triptan therapy for migraine: number needed to treat and relative cost to achieve relief within 2 hours
|Adelman J U, Belsey J
This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn.
Six drugs from the triptan class (5-HT1B/1D agonists) with different dosage were compared for the acute treatment of migraine. The drugs studied were sumatriptan (50 and 100 mg), zolmitriptan (2.5 and 5 mg), naratriptan (2.5 mg), rizatriptan (10 mg), almotriptan (12.5 mg), and frovatriptan (2.5 mg).
Economic study type
The study population comprised patients with acute migraine from the randomised trials of oral triptans included in the meta-analysis.
The setting was primary and secondary care. The economic study was carried out in the USA.
Dates to which data relate
In relation to the effectiveness evidence, the time period searched for randomised controlled trials (RCTs) was January 1990 to February 2002. The cost per dose of each triptan was taken from 2002.
Source of effectiveness data
The effectiveness data were derived from completed studies.
Link between effectiveness and cost data
The costing was undertaken retrospectively after the effectiveness results were known.
Outcomes assessed in the review
The clinical end point used was "pain free at 2 hours" as the indicator of efficacy. This is the end point currently recommended by the International Headache Society. It is also the end point that patients identify as being the "most important", and it correlates well with return to full function. Therefore, the outcome assessed was the percentage of patients who were pain free within 2 hours of the initial dosing.
Study designs and other criteria for inclusion in the review
RCTs were included. The inclusion criteria were as follows:
randomised, double-blind trial with a placebo-controlled arm;
single-dose triptan treatment, with no rescue medications or repeat doses of triptan allowed for 2 hours after the initial dosing;
data from a standard 4-point assessment scale were available for baseline and post-treatment headache severity analysis; and
definitive data were available for determining the percentage of patients (in both the treatment and control arms) who were pain free at 2 hours post dose.
Studies were excluded from the meta-analysis on the grounds of lack of relevance. More specifically, not randomised, open label, end points solely pharmacological, use of end points other than pain free, and pain relief used. Studies were also excluded if they used a drug, dose or formulation not approved in the USA.
Data from 8 additional treatment arms from included studies were not incorporated in the meta-analysis because they related to doses below that recommended for most patients (sumatriptan 25 mg, naratriptan 1 mg, rizatriptan 5 mg, and almotriptan 6.25 mg). No data for naratriptan 1 mg were included because this is only indicated in a prophylactic role.
Sources searched to identify primary studies
MEDLINE and EMBASE were searched electronically. The reference lists from primary or benchmark papers and review articles were searched manually.
Criteria used to ensure the validity of primary studies
Methods used to judge relevance and validity, and for extracting data
The methods used to judge relevance, validity or extracting data were not reported. However, the authors stated that, although direct comparative trials are the ideal means of comparing treatments, only a few studies that involved direct, head-to-head comparisons of triptans were found. Nonetheless, nearly all triptan trials used comparable protocols. For the sake of analysis, it was therefore assumed that all placebo-controlled studies not involving direct comparison of triptans were fundamentally comparable.
Number of primary studies included
Forty-five RCTs were identified in the primary search. An additional 4 studies were identified in the frovatriptan data summary. Twenty-seven of these studies, incorporating 36 active treatment arms, qualified for inclusion in the meta-analysis.
Methods of combining primary studies
The primary studies were combined using a meta-analysis. Combined data for the number-needed-to-treat (NNT) were derived using the random-effects model of DerSimonian and Laird.
Investigation of differences between primary studies
Results of the review
The percentage of patients who were pain free within 2 hours after drug administration ranged from 11.6% for frovatriptan 2.5 mg to 40.8% for rizatriptan 10 mg.
The placebo response rates ranged from 2.6% in frovatriptan 2.5 mg studies to 14.4% in almotriptan 12.5 mg studies.
The absolute percentage of patients who were pain free at 2 hours was significantly higher for all triptan doses than for placebo, (p<0.0001).
The NNT values ranged from 3.2 for rizatriptan 10 mg to 11.3 for frovatriptan 2.5 mg.
For sumatriptan 50 mg, the percentage of pain-free patients was 24.6% for the active arm and 5.8% for the placebo arm. The NNT was 5.4 (95% confidence interval, CI: 4.3 - 9.4; p<0.01).
For sumatriptan 100 mg, the percentage of pain-free patients was 30.4% for the active arm and 7.8% for the placebo arm. The NNT was 4.7 (95% CI: 4.0 - 5.9; p<0.05).
For rizatriptan 10 mg, the percentage of pain-free patients was 40.8% for the active arm and 7.8% for the placebo arm. The NNT was 3.2 (95% CI: 2.9 - 3.5; p<0.001).
For zolmitriptan 2.5 mg, the percentage of pain-free patients was 29.2% for the active arm and 8.6% for the placebo arm. The NNT was 5.1 (95% CI: 3.7 - 8.2; p<0.05).
For zolmitriptan 5 mg, the percentage of pain-free patients was 31.7% for the active arm and 6.0% for the placebo arm. The NNT was 4.2 (95% CI: 2.9 - 7.5; p<0.05).
For naratriptan 2.5 mg, the percentage of pain-free patients was 20.7% for the active arm and 8.4% for the placebo arm. The NNT was 8.2 (95% CI: 5.0 - 21.4; p<0.001).
For almotriptan 12.5 mg, the percentage of pain-free patients was 36.0% for the active arm and 14.4% for the placebo arm. The NNT was 4.7 (95% CI: 3.5 - 7.0; p<0.01).
For frovatriptan 2.5 mg, the percentage of pain-free patients was 11.6% for the active arm and 2.6% for the placebo arm. The NNT was 11.3 (95% CI: 9.3 - 14.3; p<0.0001).
Rizatriptan 10 mg was significantly more effective than all the other triptans except zolmitriptan 5 mg. Although the hierarchy of clinical effectiveness of the triptans was maintained, broader overlaps in the CIs were obtained with the NNT calculation.
Measure of benefits used in the economic analysis
The measure of benefit used for the synthesis with the costs was the number of patients with freedom of pain within 2 hours of the initial dosing. However, the NNT was calculated as the reciprocal of the measure of benefit and was multiplied by the costs in the cost-effectiveness analysis. The NNT was derived from the results of the review.
The only direct cost included was that of prescription drugs. The costs were not discounted as the time horizon was only 2 hours. The only costs considered were those for a single triptan dose since the measure of benefit was defined as freedom of pain within 2 hours. Other quantities and costs (e.g. physician visits and rescue medication) were not analysed. Dosage prices were taken from a retail website. These were lower than those at community pharmacies but were higher than the discounted prices, before member co-payment, available to many managed care organisations. The price year was 2002.
Statistical analysis of costs
The costs were treated in deterministically.
No indirect costs were included.
No sensitivity analysis was reported.
Estimated benefits used in the economic analysis
See the 'Effectiveness Results' section.
The price of a single dose of each agent was:
$10.33 for almotriptan 12.5 mg;
$13.44 for zolmitriptan 2.5 mg;
$14.41 for frovatriptan 2.5 mg;
$14.97 for sumatriptan 50 mg and $14.97 for sumatriptan 100 mg;
$15.24 for rizatriptan 10 mg;
$15.68 for zolmitriptan 5 mg; and
$17.32 for naratriptan 2.5 mg.
Synthesis of costs and benefits
The mean cost to achieve pain-free status in one patient within 2 hours post dose was:
$48.34 for rizatriptan 10 mg;
$48.57 for almotriptan 12.5 mg;
$65.18 for zolmitriptan 5 mg;
$70.83 for sumatriptan 100 mg;
$75.67 for sumatriptan 50 mg;
$78.74 for zolmitriptan 2.5 mg;
$141.43 for naratriptan 2.5 mg; and
$162.49 for frovatriptan 2.5 mg.
A closer analysis of the cost-effectiveness data revealed that, although both rizatriptan 10 mg and almotriptan 12.5 mg had a similar cost-effectiveness ratio ($48.34 versus $48.57), this did not necessarily imply that the two agents were comparable. In the case of rizatriptan, this cost-effectiveness ratio reflected high levels of clinical efficacy (NNT=3.2), coupled with an average price ($15.24 per dose). For almotriptan, average efficacy (NNT=4.7) was coupled with a low price per dose ($10.33) to achieve the same result. At the other end of the scale, lower efficacy and average, or above average, pricing resulted in much higher cost-effectiveness ratios for naratriptan 2.5 mg ($141.43) and frovatriptan 2.5 mg ($162.49).
The methods used were unclear. However, the authors conducted statistical tests on these results compared with rizatriptan, which were all highly significant.
The results of the meta-analysis indicated that there were differences in the ability of individual oral triptans to completely relieve migraine pain within 2 hours, in spite of the fact that triptans are generally considered to be equally effective. Rizatriptan 10 mg and zolmitriptan 5 mg were the most clinically effective triptans based on their measure of clinical effectiveness. When clinical effectiveness was computed with medication costs, rizatriptan 10 mg emerged as the most cost-effective triptan, together with almotriptan 12.5 mg.
CRD COMMENTARY - Selection of comparators
The choice of the comparators was explicitly justified. The justification given was that the most recent advance in the acute treatment of moderate to severe migraine has been the introduction of the 5-HT1B/1D agonists, a migraine-specific class of drugs known as triptans. You should judge whether these drugs are relevant in your own setting, or whether other comparators from other drug classes could also have been relevant.
Validity of estimate of measure of effectiveness
The authors stated that a systematic review of the literature was undertaken. The method and conduct of the systematic review were reported. The effectiveness estimates were combined in a meta-analysis and the estimates were derived credibly from the primary studies. The authors reported the methods used to derive estimates of effectiveness (NNT) and justified their choice of assumptions with reference to the medical literature. These estimates were not investigated in sensitivity analyses
Validity of estimate of measure of benefit
The measure of health benefit was proxied directly by a single effectiveness estimate. The choice was justified in terms of a recommendation from the literature, as it incorporated the features that patients wanted from treatment. The measure chosen was context specific and can only be compared with other migraine studies (and not with other economic evaluations).
Validity of estimate of costs
The authors reported that the costs were estimated from the perspective of a third-party payer. Therefore, the indirect costs were, appropriately, not included. Some costs have been omitted from the analysis (e.g. physician visits, emergency room visits and rescue medication), which might have affected the authors' conclusions. To estimate the direct costs, the authors only considered the drug acquisition cost taken from one online drug store, whose values were different from those at community pharmacies, and from the discounted prices of managed care organisations.
One of the limitations mentioned by the authors concerned the generalisability of the results to a larger population. The studies included in the review were all efficacy studies. The authors also mentioned that the study did not include the adverse effects associated with triptans in computing the NNT and cost-effectiveness. In addition, the costs associated with treating adverse effects could contribute to the overall costs of a particular treatment and, ideally, should be computed in the cost-effective calculations.
Implications of the study
From a population health perspective, the lower acquisition cost of almotriptan 12.5 mg may allow for the effective treatment of more patients than rizatriptan 10 mg for no additional medication cost. The authors suggested permitting access to triptans, and encouraged the prescribing of the more cost-effective drugs within this class.
Source of funding
Funded in part by an unrestricted educational grant from Merck & Co Inc., West Point (PA), USA.
Adelman J U, Belsey J. Meta-analysis of oral triptan therapy for migraine: number needed to treat and relative cost to achieve relief within 2 hours. Journal of Managed Care Pharmacy 2003; 9(1): 45-52
Other publications of related interest
Belsey J. The clinical and financial impact of oral triptans - an updated meta-analysis. Journal of Medical Economics 2002;5:79-89.
Edmeads J, Mackell JA. The economic impact of migraine: an analysis of direct and indirect costs. Headache 2002;42:501-9.
Ferrari MD, Roon KI, Lipton RB, et al. Oral triptans (serotonin 5-HT1B/1D agonists) in acute migraine treatment: a meta-analysis of 53 trials. Lancet 2001;358:1668-75.
Rapoport AM, Adelman JU. Cost of migraine management: a pharmacoeconomic overview. American Journal of Managed Care 1998;4:531-45.
Ryan R, Geraud G, Goldstein J, et al. Clinical efficacy of frovatriptan: placebo-controlled studies. Headache 2002;42 Suppl 2:S84-S92.
Subject indexing assigned by NLM
Administration, Oral; Cost-Benefit Analysis; Dose-Response Relationship, Drug; Female; Humans; Male; Migraine Disorders /drug therapy; Randomized Controlled Trials as Topic; Serotonin Receptor Agonists /administration & Sumatriptan /administration & Triazoles /administration & Tryptamines; dosage /economics /therapeutic use; dosage /economics /therapeutic use; dosage /economics /therapeutic use
Date bibliographic record published
Date abstract record published